The Properties tab of a project contains both system-defined properties that you cannot modify, and project specific properties that can be modified, such as the description of the project. None of the project properties are required, except for the name and location.
To set the project properties, click the Properties tab, modify the property values, and then click .
Table 4.1 List of Project Properties
Property Description
Name Specifies the name of the project. A
project can be renamed only from the Projects category view.
Description Specifies the description of the project.
Model function Specifies the type of output that your predictive model project generates.
Ensure that the types of models that you are going to use in the project fit within the selected model function type. For more information, see Table 4.2 on page 34.
Operational status Specifies the current state of the project and its models:
Prototype
indicates that the project has started, but that the project’s models have not yet been deployed. This is the default status when you create a new project.
Note: This status replaces the previous status value of “Under development”.
Deployed
indicates that at least one model within the current project has been
successfully published to a destination.
In production
indicates that the project contains models that have been successfully deployed and are being used in a production environment. This status must be set manually.
Note: The status replaces the previous status value of “Active”.
Property Description Deactivated
indicates that the project contains one or more models that were previously in a production environment and have
Location Specifies the location of the project in
the common model repository.
Champion version Specifies the project version that contains the champion model.
Champion model name Specifies the name of the model that is set as the project champion.
UUID Specifies the universally unique identifier
for a project object.
External URL Specifies a user-defined URL to a
project object in another application or to documentation related to the project.
External project ID Specifies the project ID for a project that was registered from an external
application, such as Model Studio.
Model Comparison and Evaluation This section contains project properties that are used for comparing, assessing, and evaluating the health of models.
Default training table Specifies the default training table, which can be used to test models on-demand or to validate models after a user publishes them to a publishing destination. This property is optional.
Target variable Specifies the name of the target variable that was used to train the model.
Target event value Specifies the target variable value that defines the desired target variable event.
Target values For class, nominal, ordinal, or interval targets, the set of possible outcome classes, separated by commas. For example, binary class target values might be 1, 0 or Yes, No. Nominal class target values might be Low, Medium,
Property Description
High. These values are for information only.
Target level Specifies the target level of binary, nominal, ordinal, or interval.
Output category variable Specifies the output variable name, when the Model function property is set to Text categories.
Output event probability variable Specifies the output event probability variable name, when the Model function property is set to
Classification, Analytical, Computer vision - Image classification, or Transformation.
Note: This property is also displayed for user-defined model functions.
Output prediction variable The output prediction variable name, when the Model function property is set to Prediction, Analytical, Computer
Note: This property is also displayed for user-defined model functions.
Output segmentation variable The output segmentation variable name, when the Model function property is set to Clustering, Analytical,
Segmentation or Transformation.
Note: This property is also displayed for user-defined model functions.
Output sentiment variable Specifies the output variable name, when the Model function property is set to Text sentiment.
Output topic variable Specifies the output variable name, when the Model function property is set to Text topics.
Model Comparison Criteria This section contains the properties that are used for comparing and assessing models.
Assessment criteria Specifies performance indicators that are used to assess a model. The applicable
Property Description
indicators are displayed in the drop-down list based on the specified model function.
Test condition Specifies the operator to use with the threshold value to determine whether the model is performing within an acceptable tolerance for the associated assessment indicator. An example is whether the Misclassification rate for the model is below 25%.
Threshold Specifies the threshold value to use with
the test condition operator to determine whether the model is performing within an acceptable tolerance for the associated assessment indicator.
Table 4.2 Types of Model Functions
Model Function Description
Analytical Function for any model that is not
Prediction, Classification, Clustering, or Segmentation.
Classification Function for models that have target variables that contain binary, categorical, or ordinal values.
Clustering Function for clustering models.
Computer vision - Image classification Function for computer vision models that are used for image classification.
Computer vision - Object detection Function for computer vision models that are used for object detection in a classification network.
Computer vision - Keywords detection Function for computer vision models that are used for keywords detection in neural networks.
Computer vision - Image segmentation Function for computer vision models that are used for image segmentation.
Prediction Function for models that have interval
targets with continuous values.
Segmentation Function for segmentation models.
Model Function Description
Single-series forecasting Function for models used to forecast future data based on past data for a single time series.
Text analytics Function for a SAS Visual Text Analytics modeling project.
Note: This model function appears on the Properties tab of a project only when a SAS Visual Text Analytics project contains different types of text models. It is not displayed as an option in the model function drop-down list Text categories Function for SAS Viya Text Analytics
categorization models.
Text concepts Function for SAS Viya Text Analytics
concepts models.
Text sentiment Function for SAS Viya Text Analytics sentiment models.
Text topics Function for SAS Viya Text Analytics
topics models.
Transformation Function for models that are used to determine mathematical functions, which can be used to stabilize variances, remove nonlinearity, and correct non-normality in variables to improve the fit of your model.